DEVELOPMENT OF A MODEL CONTROL STRATEGY FOR PREDICTING THE GRANULATION PROCESS IN AMMONIA-UREA PRODUCTION

This paper presents a Model Predictive Control (MPC) strategy for a multivariable, strongly coupled, and time-delay continuous granulation process in ammonia–urea production. The study focuses on the granulation stage, where product quality is primarily determined by bulk density and particle size distribution. An asymmetric control formulation is proposed, in which bulk density is required to track a predefined reference trajectory, while particle size indicators are maintained within specified bounds rather than fixed setpoints. A linear discrete-time model with input–output delays is employed to represent process dynamics. At each sampling instant, a constrained quadratic optimization problem is solved to compute optimal control actions. The performance of the proposed MPC approach is evaluated through simulation and compared with conventional PID control. The results demonstrate that MPC significantly improves tracking performance, reduces overshoot, and shortens settling time. In particular, the Integral Squared Error (ISE) is
reduced by approximately 70-75%, while overshoot is limited to 2–3%, which is considerably lower than that achieved with PID control. Moreover, MPC effectively enforces constraints on particle size distribution, whereas PID control leads to noticeable constraint violations. Overall, the proposed control strategy provides a robust and efficient solution for improving product quality, maximizing yield, and enhancing operational performance in industrial granulation systems.